Silver-Palladium Surfaces Inhibit Biofilm Formation

Abstract

Undesired biofilm formation is a major concern in many areas. In the present study, we investigated biofilm-inhibiting properties of a silver-palladium surface that kills bacteria by generating microelectric fields and electrochemical redox processes. For evaluation of the biofilm inhibition efficacy and study of the biofilm inhibition mechanism, the silver-sensitive Escherichia coli J53 and the silver-resistant E. coli J53[pMG101] strains were used as model organisms, and batch and flow chamber setups were used as model systems. In the case of the silver-sensitive strain, the silver-palladium surfaces killed the bacteria and prevented biofilm formation under conditions of low or high bacterial load. In the case of the silver-resistant strain, the silver-palladium surfaces killed surface-associated bacteria and prevented biofilm formation under conditions of low bacterial load, whereas under conditions of high bacterial load, biofilm formation occurred upon a layer of surface-associated dead bacteria.

Undesired biofilm formation is a major concern in many areas, such as medical settings, water distribution systems, and the food industry. Bacteria in biofilms are more tolerant to disinfecting operations and antibiotic therapies than planktonic bacteria, making these treatments less effective or ineffective (5, 6, 7, 8, 19, 24, 28). Biofilm formation on medical implants causes significant problems, and currently the only effective method for curing implant-associated biofilm infections involves replacement of the implant (8, 18). Biofilms in food-processing plants and in water distribution systems may harbor pathogens, causing hygienic risks, and may cause other adverse effects, such as material corrosion (9, 15, 19, 26, 28).

It is of high priority to develop methods or compounds for combating biofilms. Small molecules that may affect the bacteria and inhibit critical steps in biofilm formation as well as different surface coatings that may inhibit biofilm formation are among the strategies that have been actively pursued (10, 11). Different kinds of physical treatments have also been investigated as potential means of inhibiting biofilm formation. For example, recent studies have shown that biofilm formation can be inhibited by applying an electric field or current on a particular surface (6, 27). It has also been reported that the efficacy of antibiotics against biofilm bacteria can be increased if the antibiotics are given in combination with an applied electric current (6, 27). However, in many areas, applying electric potential or current on a surface is not feasible.

In the present report we investigate biofilm-inhibiting properties of a silver-palladium (Ag-Pd) surface. The Ag-Pd surface has been described previously (20, 21), and preliminary results of thermodynamic calculations, electrochemical tests, and antimicrobial activity have been published (2, 3). The design of the Ag-Pd surface is based on Ag upon which Pd is incompletely deposited as a microhole-structured layer, partially exposing Ag through the microholes (21). Due to the potential difference between Ag and Pd (200 mV in water) (21), Ag and Pd on the surface can be regarded as two discrete electrodes (anode and cathode), and the surface can have numerous discrete anodic and cathodic areas, generating numerous microelectric fields that may kill bacteria that approach the surface. The designed distance between Ag and Pd is less than 5 μm to ensure a high local strength of the microelectric fields because a potential difference over a short distance can give high field strengths (∼100 mV/μm). In addition to the effects of the microelectric fields, the Ag-Pd surface can also kill bacteria via redox processes. Some Ag can react to form silver chloride (AgCl) during Pd deposition. Ag ions or AgCl can be reduced to Ag by oxidation of hydroxyl groups on organic species such as surface molecules on bacteria (21, 25). After the oxidation, AgCl can be regenerated in the presence of oxygen (under aerobic conditions), where Ag is oxidized in connection to oxygen reduction on Pd. These back-and-forth redox processes of Ag converting to AgCl and AgCl converting to Ag can continuously happen if hydroxyl groups on bacteria reach an Ag-Pd surface and undergo oxidation.

Here, we report experiments with silver-sensitive and silver-resistant E. coli strains which demonstrate that Ag-Pd surfaces can inhibit biofilm formation by killing the bacteria. Experiments in batch and flowthrough systems provide evidence that silver-sensitive bacteria are killed due to a combination of microelectric fields/redox processes on the Ag-Pd surface and release of toxic levels of Ag⁺ from the Ag-Pd surface, whereas silver-resistant bacteria are killed due to microelectric fields/redox processes on the Ag-Pd surface. In addition, our experiments demonstrate an inherent weakness of antimicrobial surfaces, namely, that they allow biofilm formation upon a conditioning layer under some conditions.

MATERIALS AND METHODS

Bacteria and growth conditions.

Escherichia coli J53 (13) and E. coli J53[pMG101] (12) were used as the silver-sensitive and silver-resistant model organisms, respectively, in this study. Batch cultivation of E. coli was carried out at 37°C in AB minimal medium (4) supplemented with glucose (6.25 g/liter), methionine (25 mg/liter), proline (25 mg/liter), and thiamine (2.5 mg/liter). Flow chamber cultivation of E. coli was carried out at 37°C in FAB medium (14) supplemented with glucose (0.125 g/liter), methionine (2 mg/liter), proline (2 mg/liter), and thiamine (0.2 mg/liter).

Coupon preparation.

To obtain Ag-Pd coupons, Ag (99.9% Ag) plates were treated by immersion plating in palladium chloride solution, which was prepared from 0.5 g/liter PdCl₂ and 4/g liter NaCl dissolved in water. The coupons of stainless steel grade AISI 316L (approximately 68.7% Fe, 16.9% Cr, 10.16% Ni, and 2.02% Mo) and Ag (99.9% Ag) were used as controls. The sizes of the coupons of Ag-Pd, steel, and Ag were 4 mm by 7 mm by 0.5 mm (for batch assays) and 2 mm by 14 mm by 0.5 mm (for flow chamber assays).

Cultivation of biofilms in batch assays.

The batch assays for biofilm cultivation were performed in multiwell dishes. Each coupon was placed in a well of a multiwell dish; 5 ml of E. coli overnight cultures diluted 100-fold was transferred to each well, and the multiwell plates were incubated at 37°C with shaking at 60 rpm for 72 h. Prior to microscopic investigation, the spent medium was removed from the wells, and fresh AB minimal medium was added, after which Live/Dead stain was added as described below. For the determination of the number of CFU in the 72-h multiwell cultures, vigorously vortexed serial dilutions of cell suspensions were plated on LB (1) agar plates, and colonies were counted after 30 h of incubation at 37°C.

Cultivation of biofilms in continuous flow chamber assays.

The flow chamber systems for biofilm cultivation were assembled and prepared as described previously (23). Each coupon was installed in a flow chamber that was subsequently inoculated by injecting 250 μl E. coli overnight culture diluted 100-fold using a small syringe. After inoculation, adhesion of cells to the coupon surfaces was allowed for 1 h without flow, and afterwards FAB medium was started to flow through the chambers at a mean flow velocity of 0.2 mm/s, corresponding to laminar flow with a Reynolds number of 0.02, using a Watson Marlow 205S peristaltic pump (Watson Marlow, United Kingdom). Biofilms were investigated microscopically after 24 and 72 h.

Microscopy and image acquisition.

Biofilms on the coupon surfaces were observed by the use of a Zeiss LSM 510 META (Carl Zeiss, Germany) confocal laser scanning microscope (CLSM) and staining with the Live/Dead BacLight Bacterial Viability Assay (Invitrogen), which utilizes green fluorescent SYTO 9 (Invitrogen) for staining of cells (5 μM for batch-grown biofilms and 5 μM for flow chamber-grown biofilms), and red fluorescent propidium iodide (Sigma, Germany) for staining of membrane-compromised cells (40 μM for batch-grown biofilms and 20 μM for flow-chamber-grown biofilms). Although we cannot exclude that some propidium iodide-stained cells were membrane compromised but not dead, we have assumed in the following discussion that all propidium iodide-stained cells were dead. Images were obtained using a 63× objective with a 0.95 numerical aperture for batch assays and a 40× objective with a numerical aperture of 1.30 for flow chamber assays. A 488-nm argon laser was used to excite the SYTO 9-stained cells, and a 543-nm helium/neon laser was used to excite the propidium iodide-stained cells. Simulated three-dimensional images were generated by the use of IMARIS software (Bitplane, Switzerland).

RESULTS AND DISCUSSION

In order to study a potential biofilm-inhibiting effect of the Ag-Pd surface, metal coupons were placed in the wells of multiwell dishes, and diluted E. coli J53 overnight cultures were added to the wells; after incubation for 72 h, bacteria associated with the surface of the metal coupons were stained with the Live/Dead BacLight stain and visualized by the use of CLSM. Metal coupons with steel and Ag surfaces were included as controls along with the coupons with the Ag-Pd surface. As shown in Fig. 1A and B, the bacteria formed biofilms on both the steel surface and the Ag surface, the only apparent difference being that a few dead cells were present close to the Ag surface. In agreement with this finding is a report that metallic Ag has only a slight antimicrobial effect because of its chemical stability (22), and our previous studies have also demonstrated that pure Ag does not have a significant inhibiting effect on biofilm formation (16). Biofilm formation was inhibited, however, on the Ag-Pd coupons that had only a few dead bacteria scattered on the Ag-Pd surface (Fig. 1C). Because Ag-Pd surfaces were shown to release more Ag⁺ to the surrounding liquid than Ag surfaces (2, 3, 17), we could not exclude the possibility that the bacteria in the wells with the Ag-Pd coupons were killed by the high Ag⁺ levels. We therefore determined the number of live planktonic bacteria in the wells by plating serial dilutions on LB agar plates and determining the number of CFU after incubation. As shown in Fig. 2, the wells with the steel and Ag coupons contained approximately 10⁸ CFU/ml, whereas no live bacteria were found in the wells with the Ag-Pd coupons. From these experiments we therefore could not conclude whether the lack of biofilm formation on the Ag-Pd surface was due to cell killing by microelectric field/redox processes or by Ag⁺ toxicity.

FIG. 1.

CLSM micrographs of batch-grown, 72-hour-old, Live/Dead-stained E. coli J53 biofilms on steel (A), Ag (B), and Ag-Pd (C). The top view shows the biofilms from the growth medium side, whereas the bottom view shows the biofilms from the metal coupon side. Green fluorescence indicates live cells, and red fluorescence indicates dead cells. The images are representative of three independent experiments. Bar, 10 μm.

FIG. 2.

Effect of steel, Ag, and Ag-Pd coupons on the number of CFU of planktonic E. coli cultures after 72 h of batch cultivation. Means and standard deviations (error bars) of three replicates are shown. White bars indicate E. coli J53, and black bars indicate E. coli J53[pMG101].

In an attempt to separate the effects of microelectric field/redox processes from Ag⁺ toxicity, we employed the silver-resistant strain E. coli J53[pMG101]. As shown in Fig. 3, the E. coli J53[pMG101] strain formed biofilm on the steel, Ag, and Ag-Pd surfaces. However, unlike the biofilms on the steel and Ag coupons (Fig. 3A and B), the biofilms on the Ag-Pd coupons had a layer of dead cells close to the Ag-Pd surface (Fig. 3C and D). Experiments with shorter incubation times than 72 h indicated that the bacteria were killed shortly after attaching to the Ag-Pd surfaces (Fig. 3E). Determinations of the numbers of CFU showed that the wells with steel, Ag, and Ag-Pd coupons all contained approximately 10⁸ CFU/ml of the E. coli J53[pMG101] bacteria (Fig. 2). Taken together, these experiments suggested that the bacteria close to the Ag-Pd surface were killed due to the microelectric field/redox processes and that bacteria from the planktonic phase subsequently formed biofilm upon the layer of dead bacteria.

FIG. 3.

CLSM micrographs of batch-grown, 72-h-old (A, B, C, and D) or 24-h-old (E), Live/Dead-stained E. coli J53[pMG101] biofilms on steel (A), Ag (B), and Ag-Pd (C, D, and E). The top views of A, B, C, and E show the biofilms from the growth medium side, whereas the bottom views of A, B, C, and E show the biofilms from the metal coupon side. Panel D shows a side view of a biofilm with the lower cell layer closest to the Ag-Pd surface. Green fluorescence indicates live cells, and red fluorescence indicates dead cells. The images are representative of three independent experiments. Bar, 10 μm.

The number of planktonic bacteria in the wells of the multiwell dishes (approximately 10⁸ CFU/ml after 72 h of incubation) is much higher than in most relevant settings. In order to study the effects of the Ag-Pd surface on biofilm formation in a system with a lower bacterial load, we installed metal coupons in flow chambers, inoculated the flow chambers with E. coli J53 or E. coli J53[pMG101], irrigated the flow chambers with growth medium for 72 h, stained the bacteria with Live/Dead BacLight, and visualized the bacteria by the use of CLSM. As shown in Fig. 4 and Fig. 5, the E. coli J53 and E. coli J53[pMG101] strains formed biofilms on both the steel and Ag surfaces although the biofilm formed by the E. coli J53 strain on the Ag coupon contained a few dead bacteria close to the Ag surface. However, neither E. coli J53 nor E. coli J53[pMG101] could form biofilm on the Ag-Pd surfaces (Fig. 4C and Fig. 5C). The Ag-Pd coupons had only a few bacteria scattered on the surface. These experiments suggested that biofilm formation is inhibited on Ag-Pd surfaces under conditions where high numbers of bacteria from a planktonic phase cannot continuously initiate biofilm formation.

FIG. 4.

CLSM micrographs of flow chamber-grown, 72-h-old, Live/Dead-stained E. coli J53 biofilms on steel (A), Ag (B), and Ag-Pd (C). The top view shows the biofilms from the growth medium side, whereas the bottom view shows the biofilms from the metal coupon side. Green fluorescence indicates live cells, and red fluorescence indicates dead cells. The images are representative of three independent experiments. Bar, 10 μm.

FIG. 5.

CLSM micrographs of flow chamber-grown, 72-h-old, Live/Dead-stained E. coli J53[pMG101] biofilms on steel (A), Ag (B), and Ag-Pd (C). The top view shows the biofilms from the growth medium side, whereas the bottom view shows the biofilms from the metal coupon side. Green fluorescence indicates live cells, and red fluorescence indicates dead cells. The images are representative of three independent experiments. Bar, 10 μm.

Although biofilm formation on the flow chamber-installed Ag-Pd coupons was inhibited in the case of both the silver-sensitive E. coli J53 strain and the silver-resistant E. coli J53[pMG101] strain, the outcomes of the Live/Dead BacLight staining were not identical. In the case of the silver-sensitive E. coli J53 strain, all the surface-attached bacteria were red (Fig. 4C) and supposedly dead, whereas in the case of the E. coli J53[pMG101] strain, the surface-attached cells were either red or green (Fig. 5C). The fact that some green-stained, and supposedly live, E. coli J53[pMG101] cells were present on the flow chamber-installed Ag-Pd coupons might reflect the microheterogeneity of the Ag-Pd surface. The surface may contain some sites where attached E. coli J53[pMG101] bacteria can live, but the daughter cells cannot establish on the surface close by and may be shed to the planktonic phase. This implies that the silver-sensitive E. coli J53 bacteria, present on the flow chamber-installed Ag-Pd coupons, were all killed either because of the microelectric field/redox processes or a local Ag⁺ concentration that was higher than these silver-sensitive bacteria could tolerate.

In conclusion, experiments with silver-sensitive and silver-resistant E. coli strains showed that Ag-Pd surfaces could inhibit biofilm formation by killing the bacteria. Batch experiments provided evidence that biofilm formation of the silver-sensitive bacteria was inhibited on the Ag-Pd surface due to release of toxic levels of Ag⁺ in addition to the killing effects of the surface, whereas biofilm formation of the silver-resistant bacteria occurred upon a layer of surface-associated dead bacteria on the Ag-Pd coupons. Unlike the batch setup, where high numbers of silver-resistant planktonic bacteria could continuously initiate biofilm formation, the flow chamber system had a lower bacterial load, and in this system the Ag-Pd surfaces proved efficient in preventing biofilm formation by both silver-sensitive and silver-resistant bacteria. We envision that it may be beneficial to coat, for example, the vulnerable parts of medical implants, medical equipment, water distribution systems, or food production facilities with biofilm-inhibiting Ag-Pd surfaces. However, as biofilm formation evidently can occur if the antimicrobial surface becomes covered with a conditioning layer, the highest efficiency of an Ag-Pd surface would be achieved under conditions where appropriate cleaning practices can be applied.